Intercom Fin
The #1 AI Agent for customer service — top-ranked on bake-offs and complex queries.
Acquired Intercom Fin
Acquired · June 2026
Fin is the AI customer-service agent built by Intercom, which renamed itself Fin as the agent became its flagship — resolving roughly three-quarters of support volume end-to-end across chat, email, WhatsApp, SMS, phone and Slack. On June 15, 2026 Salesforce announced a ~$3.6 billion agreement to acquire the company and fold Fin into Agentforce. The product is still operating, with the deal expected to close in Salesforce’s fiscal Q4 2027.
Acquired by Salesforce.
Last known pricing: Usage-based (per resolution) + paid plans
Is Intercom Fin shut down?
No — Intercom Fin is still operating, but it no longer stands alone: it was acquired by Salesforce in Jun 2026 and continues under the acquirer's ownership. We list it in the ToolDirectory.AI graveyard's acquired-but-operating section and watch for post-acquisition changes.
What to use instead
Decagon
AI customer support concierge that resolves real tickets across chat, email, and voice.
Paid - Inquire
4.92
460
Sierra AI
AI-powered platform for enhancing customer experience through conversational AI.
Paid - Inquire
4.93
510
Featured
Cresta AI
Generative AI for the intelligent contact center, offering self-service, live coaching, and post-call insights.
Paid - Inquire
4.8
300
Overview
Intercom Fin: The #1 AI Agent for Customer Service
Fin is Intercom's AI Agent for customer service — and ranks #1 in head-to-head bake-offs, on benchmarks, and on complex queries. Fin handles the full range of customer support: answering questions, resolving issues end-to-end, and escalating cleanly to human agents when needed. Embedded in Intercom's messenger and ticketing platform, Fin is the path of least resistance for any team already on Intercom.
Fin is built on the latest frontier models with Intercom's purpose-built fine-tuning for customer-service quality and safety. It can be deployed in minutes on top of your existing knowledge base and ramps to deflection rates that materially change support cost structures.
Acquisition note (June 2026): On June 15, 2026, Salesforce announced a definitive agreement to acquire Fin (the AI agent from Intercom, which renamed itself Fin) for approximately $3.6 billion, bringing its customer-service AI agent to Salesforce Agentforce. The deal is expected to close in Salesforce’s fiscal Q4 2027, and Fin continues to operate normally in the meantime.
Key Features:
- #1 ranked AI agent on bake-offs and complex queries
- End-to-end resolution, not just deflection
- Native to Intercom Messenger and Inbox
- Connects to your knowledge base out of the box
- Multi-language support
- Custom actions and integrations
- Smooth human-in-the-loop handoff
- Powered by frontier LLMs with custom fine-tuning
- Production deployments in days, not months
Ideal Use Case:
Intercom Fin is ideal for support orgs that want to deploy a top-tier AI agent without building from scratch. Especially strong for teams already on Intercom — Fin sits natively in the inbox and messenger and connects to existing knowledge bases in minutes.
Why Use Fin:
- #1 AI agent ranking by independent bake-offs
- Native to Intercom — fastest deployment
- End-to-end resolution, not just answer-suggesting
- Multi-language and global out of the box
- Frontier models with support-specific fine-tuning
FAQ
Do I need Intercom to use Fin? Fin is part of the Intercom platform — best fit if you are on Intercom or willing to migrate. How is Fin different from generic AI chatbots? Fin is benchmarked end-to-end on customer service workflows, not just deflection metrics. What does Fin connect to? Knowledge bases, custom actions, your CRM, internal tools. Multi-language? Yes — production-ready across major languages.
Q: Who owns Intercom Fin? A: Salesforce — on June 15, 2026 it announced a definitive agreement to acquire Fin (formerly Intercom) for about $3.6 billion. Fin’s AI customer-service agent continues to operate and is being brought into Salesforce Agentforce, with the deal expected to close in Salesforce’s fiscal Q4 2027.
tl;dr:
Intercom Fin is the #1-ranked AI Agent for customer service — bake-off winner, native to Intercom, end-to-end resolution. Built on frontier models with support-specific fine-tuning.
Related
Looking for more options? Browse the Customer Support directory or read our best AI customer support tools listicle. Intercom Fin is also tracked on Crunchbase.
Why Use Intercom Fin
FAQ
Editorial Review
Our take on Intercom Fin.
Intercom Fin is an AI agent for customer service that handles complex support queries with high accuracy in comparative evaluations.
What works
- High accuracy on complex queries in benchmarking
- Integrates with Intercom's support platform
- Strong community rating reflects real-world reliability
What doesn't
- Vendor lock-in if you're not already on Intercom
- Pricing opacity requires sales conversations
Intercom Fin functions as a customer service AI agent designed to resolve support tickets and handle inquiries without human intervention. According to the supplied facts, it ranks highly in comparative bake-offs and performs well on complex queries—the kinds of requests that typically require judgment calls or multi-step reasoning. The tool sits within Intercom's broader platform, which means it integrates with existing ticketing, customer data, and routing workflows if you're already using their suite.
What matters most here is whether Fin actually closes tickets or just looks good in benchmarks. The community rating of 4.91 suggests real users find it reliable in practice, not just in controlled tests. That said, customer service AI is crowded: Decagon, Sierra AI, and Cresta AI all compete in the same space, each with different philosophies about how much human judgment to preserve. Fin's strength appears to be accuracy on difficult cases, which matters more than speed if your support team spends time fixing AI mistakes.
The main trade-off is lock-in. If you're evaluating Fin, you're probably considering it within Intercom's ecosystem. That's fine if you already use them; it's a cost to weigh if you don't. Pricing requires a conversation with their sales team, so budget-conscious teams should factor in the discovery phase.
